A solution of obstacle collision avoidance for robotic fish based on fuzzy systems

This paper is addressed in fuzzy decision made to change on trajectory direction of robot´s path. The generation of membership functions for fuzzy systems is a challenging problem. While robotic fish is swimming in the environment with a high potential of hurdles, it is possible for them to meet with collisions because there are many kinds of obstructions in water. Therefore, approaching natural and smooth movements for robotic fish is related to the detecting and recognizing obstacles as well as trying to avoid any kind of collision. Two fundamental data are the measuring of distance from sensor to obstacles and the possible existence of obstacles. Because the data is nonlinear, they can be solved with the fuzzy trajectory direction. The changing direction of trajectory of a robot should be made so that the robot can move in the direction with no hurdles. The experimental results indicate that robotic fish changes the trajectory planed following the proposed fuzzy decision results within the higher obstacles density without collision.